Abstract: How do export demand shocks affect firms' product portfolios, and which firms respond by adopting new products? We study these questions using Vietnamese manufacturing firms during the US-China trade war, which generated a positive export demand shock for Vietnamese producers of goods subject to US tariffs on Chinese exports. Combining administrative firm level data with a measure of each firm's latent advantage in products outside its current portfolio, we find that firms are more likely to begin producing tariff-exposed products, and that this response is stronger for firms with greater latent advantage in those products. Our results indicate that latent advantage, in addition to revealed advantage in existing products, determines which firms respond to new export opportunities. These findings have implications for industrial policies that promote the production of goods with high global market potential.

Abstract: Not-for-profit organisations are frontline providers when disasters strike, yet how disasters affect them is largely unknown. Using Australian administrative data, we estimate the effects of the January 2011 Brisbane flood over the following six years, comparing inundated organisations with non-inundated neighbours in the same flooded areas in a difference-in-differences design. On average, survival is unaffected and non-capital expenditure falls by about 11%. However, responses differ sharply by type. Religious organisations, which hold the most capital relative to income and depend most on donations, raise capital expenditure by about 51% while cutting non-capital and labour expenditure by 43% and 61%. Their revenue from goods and services, which excludes donations, falls by about half. Community groups and social, health, and emergency services show no such contraction. Because few organisations close, the flood reduces the activity of surviving organisations, so disaster support should fund operations during recovery as well as rebuilding.

Presented at: the 13th World Congress of the Econometric Society,  the 2024 Econometric Society Australasia Meeting,  the 2024 Monash Environmental Economics Workshop, the 2024 Asia Meeting of the Econometric Society, the Monash Green Labs Brown Bag Seminar, the 10th Singapore Economic Review Conference, the 4th Australian Workshop on Public Finance.

Abstract: Pandemics, natural disasters, and other large disruptions can prevent firms from operating, yet little is known about how much revenue is affected, whether revenue losses fade, or which firms withstand and recover best. Using administrative records covering nearly all Australian businesses, we compare established firms in a state that experienced long, repeated COVID-19 lockdowns that significantly disrupted normal operations with firms in mostly unaffected states. Surprisingly, the lockdowns did not increase the probability of firm exit. In contrast, turnover fell 5% to 6% during the lockdowns and remained persistently 6% below the control-state path in the three years after restrictions ended, about AUD 44,000 per firm per quarter. During the lockdowns, multi-location firms, incorporated firms, larger firms, and more productive firms lost less. Afterwards, only the location advantage persisted. The firms that lost the most had the weakest pre-pandemic management practices, particularly performance monitoring, while the location advantage aligns with shifting activity across sites.

Presented at: the Asia-Pacific Industrial Organization Conference 2025, the 16th Australasian Public Choice Conference, the 4th PhD Summer Institute in International Trade & Industrial Organisation.

Abstract: Governments increasingly rely on choice and competition to deliver aged care. These reforms can make prices transparent while care quality stays undisclosed. Little is known about provider behaviour under such disclosure. Using an administrative panel of 48 Australian metropolitan markets over 2008-2019 and a difference-in-differences design, we examine how residential aged care markets with higher and lower pre-reform competition evolved after accommodation prices became publicly comparable in 2014. A one standard deviation increase in pre-reform competition is associated with 4.7 per cent fewer providers in the post-reform period but only 1.8 per cent fewer facilities, so facilities per provider rose. The decline was concentrated among for-profit providers, while surviving for-profit incumbents expanded capacity and diversified into home care. Not-for-profit providers showed no differential response on any margin. The reform set out to strengthen consumer choice, yet the number of independent providers fell most in the markets where consumers had alternatives to compare. Transparency without quality disclosure may therefore consolidate care markets.

Abstract: We use rich administrative data on all manufacturing firms in Vietnam to estimate the causal effects of internal migration on firm performance. Leveraging population census microdata and a shift-share instrument based on international agricultural price shocks, we provide the first evidence on how migration affects quantity-based total factor productivity (TFPQ), as well as other firm outcomes. In the short run, migration reduces marginal costs, prices, and wages, but has no effect on TFPQ, revenue per worker, or capital intensity, and no lasting effects. Adjustment occurs through new firm entry, particularly by small firms in high-productivity sectors observable in our detailed data.

Presented at: the Australasian Development Economics Workshop 2025.

Abstract: Policies that facilitate platform comparison are expected to lower prices. In two-sided markets, they can destabilize coexistence. We call the share of users able to compare platforms effective comparability. A larger share makes demand more responsive to provider participation, lowering prices but strengthening feedback between users and providers. Once the feedback loop exceeds one, symmetric coexistence becomes unstable: all comparers choose one platform, while its rival survives on captive users. The model derives this transition, shows when an induced crossing lowers welfare, and identifies the causal responses needed to measure distance from instability and distinguish comparison from contestability policy.

Abstract: Carbon leakage is typically attributed to differences in carbon prices or trade exposure. We show that institutional asymmetries can generate pollution-haven forces even when formal abatement rules are identical. We develop a model of environmental regulation with delegated enforcement, corruption, and firm mobility under asymmetric information. When intermediaries can identify relocation-prone firms, deterring intermediary-assisted exit requires raising rents for marginal types, which shifts participation and alters relocation flows. By contrast, conditional on participation, the marginal abatement rule is unaffected in the baseline model with linear costs. Embedding the model in a two-country setting, we show that relocation is environmentally neutral under symmetric institutions, but weaker enforcement abroad attracts firms and increases global emissions. Institutional quality thus emerges as a central determinant of carbon leakage.

Abstract: This paper measures the marginal impact of climate variability on Vietnamese households’ income. We combine survey data from the Viet Nam Household Living Standard Survey (VHLSS) database with daily climate data from the Climate Prediction Center to estimate the response function of Vietnamese households’ revenues to past climate variability. We focus on the non-linearity of the response and notably on the impacts of extremely warm days. We find that on average an additional day above 33°C is associated with a decrease of the yearly income by 1.3%. This strong effect (13 times higher than an equivalent study in the US case) is not specific to the agricultural sector. It is highest for the lowest deciles of the revenue distribution. Using projection scenarios under the Representation Concentration Pathways (RCP) 8.5 and 4.5, we find an estimated impact of global warming (without further adaptation) of up to 100% of households’ revenues in 2090 in some regions (Northern region and the Red River Delta area) under RCP8.5. These strong negative impacts are also likely to be specifically concentrated on poor households and to increase revenue inequalities.

Press: [The conversation] 

Presented at the GEMMES Viet Nam Seminar Series 2020/2021

Abstract: This chapter evaluates the effects of climate variability on households, individuals, and firms in Viet Nam. First, we examine the impacts of weather shocks on household income using VHLSS 2002–2018. Second, we investigate how labour supply changes with regard to climate change, using the Labour Force Survey 2010–2018. Third, we identify how temperature affects firms’ productivity, revenue, output, and size. Fourth, we evaluate the adaptation of household and individuals, and their perception of climate variability. Fifth, we provide a case study of the Mekong River Delta responding to the severe drought of 2016. Based on the results and four climate scenarios, we provide projections for losses by the end of this century. We find several notable results. First, climate variability would harm household agricultural income (from fruits and non-crop components), especially when the temperature is above 33°C. We also find that weather shocks have negative impacts on poor households in comparison to other groups. Third, we find a negative relationship between climate change and working hours/hourly wage. Fourth, temperature’s increase reduces firms’ revenue, total factor productivity, output, and size. Fifth, our results suggest the non-linear effect of temperature change on employment allocation and migration. Sixth, we find that individuals are aware of extreme weather, but unaware of gradually increasing temperatures. Finally, we find that the 2016 severe drought led to a significant increase in migration rate in Mekong River Delta